CATALOG INTELLIGENCE ENGINE
Want to back the stream? The «داعم» membership on YouTube is EGP 60 a month: badges, emoji, and I answer your questions first on stream. Become a supporter
The gist
In this April 23, 2026 recording, Luqman documents practical work on the Catalog Intelligence Engine and Command Center with agents such as Codex. He outlines a roadmap-driven workflow that schedules tasks by dependency, prepares prompts, runs batches from one terminal, audits tasks and milestones, and turns findings into repair work. He observed a faster pace, but stressed that feature completion is not product readiness: bugs, QA, UX, and Shopify requirements still needed verification, while project-root, repository-overlap, installation, and cross-environment issues remained under diagnosis or repair. Much of the practical work therefore focused on clearer README guidance, simpler setup for non-programmers, intended Windows and Linux support, and fixing problems in the product rather than burdening users. The recording also develops the product and market strategy: structured research into GEO and Shopify competitors, defining positioning and the ideal customer before redesigning the website, and differentiating through changes applied directly to merchant catalogs rather than scores and recommendations alone. Luqman considers organic content, direct outreach, and agency partnerships, arguing that genuine usage and demonstrated value should precede heavier paid acquisition. Distribution, AI-traffic attribution, pricing, and the acquisition plan remain unresolved. Audience discussions cover learning by building and conversing with AI, reducing generated-code bugs through reviewer and debugging agents, vibe-coding security, whether MCP tools transfer across agents and projects, and local models and hardware. Secondary discussions also address domain trading, vision and OCR for old Arabic manuscripts, finding product opportunities in everyday friction, Egyptian trust in digital services and subscriptions, recreating UI from visual references, screen comfort, the workload of short-form content, and fear-driven over-polishing of tools. These are presented as qualified experiences and opinions rather than definitive benchmarks; his broader advice is to focus seriously on one product, or very few, and give it a real market test instead of scattering effort.
129 moments from the stream
- 7:04:58
Milestone Planning Around Shopify QA and Submission
Luqman reviews the work sequence and expects progress to slow around quality assurance and Shopify app submission. He notes the need to review submission requirements, prepare a solid demonstration video, and establish an appropriate YouTube presence, while treating the schedule as a planning estimate rather than a confirmed outcome.
- 7:08:03
Can MCP Tools Work Across Different Agents and Projects?
Luqman examines an audience question about whether MCP files and commands are limited to particular tools or can be used by other agents when opening a different project. He restates the possible interpretations and proposes trying another agent, but the excerpt does not establish a definitive answer.
- 7:12:29
Handling Project-Path Differences Across Setups
A discussion of a missing folder when opening a project, with suggestions to add it or devise a workaround. The speaker notes that the repository setup was prepared quickly and may not suit every environment, and that overlap with an existing project root caused a central tool to crash while extracting a new repository.
- 7:14:58
Separating the Marketing Site from App UX Polish
A clarification asks whether milestone tasks concern the marketing site or the app experience inside Shopify. If the workstreams are independent, the proposal is to split them into separate milestones and place the marketing-site milestone before UX polish.
- 7:18:24
Why Completion Percentage Does Not Predict Remaining Time
A planning observation that quickly completing a large portion of a project does not imply the remainder will finish at the same pace; harder work may cluster near the end and take longer, potentially warranting additional milestones.
- 7:27:40
Planning an Admin UI and UX Polish Milestone
The speaker reviews an added milestone covering a Shopify-related admin interface, UX polish, and a marketing site, then considers when to take on the next task. He says he wants Claude to handle some UX-polish work, without establishing that it was completed or successful.
- 7:34:44
Roadmaps and Milestone Dependencies
The discussion emphasizes that a clear, well-structured roadmap is necessary to guide the work, and that tasks blocking a later milestone should be moved into the appropriate milestone so dependencies are represented accurately.
- 7:38:50
Reducing Coding Bugs with Review and Debugging Agents
In response to a question about bug-heavy generated code, the advice is to create a reviewer or debugger agent to inspect the first agent’s work, repeat the review, and diagnose each bug step by step. The speaker qualifies that this may reduce bugs but will not guarantee their disappearance, and that programming knowledge helps identify causes and guide Codex effectively.
- 7:41:43
Organizing a Multi-Agent Code-Quality Workflow
The speaker describes a setup combining specialized auditing and debugging roles, including an auditor backed by several sub-agents, as a way to help keep coding work organized and clean.
Got a question about something in this stream?
Ask in the group